Extended Data Fig. 4: Random Forest classifier in distinguishing CGIs/T2Ds from NGTs based on all metabolites. | Nature Medicine

Extended Data Fig. 4: Random Forest classifier in distinguishing CGIs/T2Ds from NGTs based on all metabolites.

From: Microbiome–metabolome dynamics associated with impaired glucose control and responses to lifestyle changes

Extended Data Fig. 4

The performance of the classifiers is assessed by area under the curve (AUC); the cross-validation AUCs based on 10-fold cross-validated repeated 10 times in the discovery cohort and external prediction AUCs in the validation cohort were provided, respectively.

Back to article page